Effective long short-term memory with fruit fly optimization algorithm for time series forecasting

Peng, Lu; Zhu, Qing; Lv, Sheng-Xiang; Wang, Lin

doi:10.1007/s00500-020-04855-2

Cited by 63 publications

(37 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are also many other approaches with successful applications in deep learning community. These neural networks are tailored for a variety of different tasks, to name a few, recurrent neural networks (RNN) [ 34 ] and long short-term memory (LSTM) [ 35 , 36 ] are suitable for modeling sequential data and sequence recognition and prediction, region-based convolutional neural network (R-CNN) [ 37 ] as well as its variants and you only look once (YOLO) [ 38 ] are capable of tackling object detection problems, SegNet [ 39 ] is tailored for semantic segmentation tasks. Such models have more complex structures and modules, such as memory blocks in LSTM, to cope with more complicated problems [ 40 , 41 ].…”

Section: Methodsmentioning

confidence: 99%

Multilabel Image Classification Based Fresh Concrete Mix Proportion Monitoring Using Improved Convolutional Neural Network

Han

Jiao

Yin

2020

Sensors

View full text Add to dashboard Cite

Proper and accurate mix proportion is deemed to be crucial for the concrete in service to implement its structural functions in a specific environment and structure. Neither existing testing methods nor previous studies have, to date, addressed the problem of real-time and full-scale monitoring of fresh concrete mix proportion during manufacturing. Green manufacturing and safety construction are hindered by such defects. In this study, a state-of-the-art method based on improved convolutional neural network multilabel image classification is presented for mix proportion monitoring. Elaborately planned, uniformly distributed, widely covered and high-quality images of concrete mixtures were collected as dataset during experiments. Four convolutional neural networks were improved or fine-tuned based on two solutions for multilabel image classification problems, since original networks are tailored for single-label multiclassification tasks, but mix proportions are determined by multiple parameters. Various metrices for effectiveness evaluation of training and testing all indicated that four improved network models showed outstanding learning and generalization ability during training and testing. The best-performing one was embedded into executable application and equipped with hardware facilities to establish fresh concrete mix proportion monitoring system. Such system was deployed to terminals and united with mechanical and weighing sensors to establish integrated intelligent sensing system. Fresh concrete mix proportion real-time and full-scale monitoring and inaccurate mix proportion sensing and warning could be achieved simply by taking pictures and feeding pictures into such sensing system instead of conducting experiments in laboratory after specimen retention.

show abstract

Section: Methodsmentioning

confidence: 99%

Multilabel Image Classification Based Fresh Concrete Mix Proportion Monitoring Using Improved Convolutional Neural Network

Han

Jiao

Yin

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…First, other public opinions on breaking news, press release, and regulatory announcements can be accessed to distinguish sentiments and extract information. Second, attempting other emerging text classification technologies, such as Graph Neural Network (GNN) and long short-term memory (LSTM) is worth doing [ [38] , [39] ]. Third, to explore the applicability of this study, other markets affected by the pandemic, such as stocks markets and gold markets, can also consider using online social information for forecasting.…”

Section: Conclusion and Future Researchmentioning

confidence: 99%

Forecasting the U.S. oil markets based on social media information during the COVID-19 pandemic

Wang

et al. 2021

Energy

Self Cite

View full text Add to dashboard Cite

Accurate oil market forecasting plays an important role in the theory and application of oil supply chain management for profit maximization and risk minimization. However, the coronavirus disease 2019 (COVID-19) has compelled governments worldwide to impose restrictions, consequently forcing the closure of most social and economic activities. The latter leads to the volatility of the oil markets and poses a huge challenge to oil market forecasting. Fortunately, the social media information can finely reflect oil market factors and exogenous factors, such as conflicts and political instability. Accordingly, this study collected vast online oil news and used convolutional neural network to extract relevant information automatically. Oil markets are divided into four categories: oil price, oil production, oil consumption, and oil inventory. A total of 16,794; 9,139; 8,314; and 8,548 news headlines were collected in four respective cases. Experimental results indicate that social media information contributes to the forecasting of oil price, oil production and oil consumption. The mean absolute percentage errors are respectively 0.0717, 0.0144 and 0.0168 for the oil price, production, and consumption prediction during the COVID-19 pandemic. Marketers must consider the impact of social media information on the oil or similar markets, especially during the COVID-19 outbreak.

show abstract

“…To ensure practicality, the delivery policy that can be used is either the traveling salesman problem policy or the vehicle routing problem policy. Additionally, exact algorithms, improved heuristics and metaheuristics (Wang et al, 2019;Peng et al, 2020) can be used to deal with these problems for satisfactory performance.…”

Section: Conclusion and Future Researchmentioning

confidence: 99%

Heuristic for the new coordinated dynamic demand lot-size and delivery planning problem

Wang

Peng

et al. 2020

Self Cite

View full text Add to dashboard Cite

Purpose The purpose of this study is to propose a new coordinated dynamic demand lot-size and delivery planning problem (CDLSDP), in which the delivery policy is integrated into the coordinated dynamic demand lot-size problem (CDLSP). Design/methodology/approach As a non-deterministic polynomial complete (NP-complete) problem, this CDLSDP seems difficult to be solved by a polynomial-time method. To handle this problem effectively and efficiently, a four-phase heuristic that balances the setup and inventory costs in the coordinating and delivery stages is designed to find near-optimal solutions. Findings Numerous computational experiments show that the proposed four-phase heuristic is effective and efficient. For 1,800 experiments with different scales, and different joint setup costs, solutions by the proposed heuristic have an average gap no more than 1.34% from the optimal solution. Research limitations/implications To decrease total system cost, the CDLSDP optimizes the time-phased replenishment and delivery schedule, which includes joint setup cost, item setup, delivery and inventory cost, for each period. An effective and efficient four-phase heuristic is designed to solve the CDLSDP. Originality/value Compared with the traditional CDLSP, the delivery policy is considered by the new CDLSDP. Moreover, the proposed four-phase heuristic is a good candidate for solving the CDLSDP.

show abstract

Effective long short-term memory with fruit fly optimization algorithm for time series forecasting

Cited by 63 publications

References 48 publications

Multilabel Image Classification Based Fresh Concrete Mix Proportion Monitoring Using Improved Convolutional Neural Network

Multilabel Image Classification Based Fresh Concrete Mix Proportion Monitoring Using Improved Convolutional Neural Network

Forecasting the U.S. oil markets based on social media information during the COVID-19 pandemic

Heuristic for the new coordinated dynamic demand lot-size and delivery planning problem

Contact Info

Product

Resources

About